ANALISIS PENGARUH PRODUK, PROMOSI, HARGA DAN TEMPAT TERHADAP KEPUTUSAN PEMBELIAN (Studi Kasus Pada Kedai Amarta Semarang )

Abstract

This study aims to determine the effect of the product, promotion, price and place and the price of purchasing decisions at Amarta Shop in Semarang. The independent variable are product, promotion, price and place affect the purchase decision as the dependent variable. The population in this study are customers Amarta Shop. Samples were taken of 100 respondents using a purposive sampling technique. Data was collected using a survey method through questionnaires filled out by customers. Then, the data obtained were analyzed using multiple regression analysis. This analysis includes the validity test, reliability test, multiple regression analysis, classic assumption test, hypothesis testing through the F test and t test, and coefficient of determination analysis (R²). Based result, obtained regression equation: Y = 0,428 X1 + 0,208 X2 + 0,018 X3 – 1,446 X4. Based on statical data analysis, the indicators in this research is valid and the variables are reliable. In testing the assumption of classical, regression model multikolonieritas, does not occur heterokedasitas, and normal distribution. Individually, the variables have greater influence is product with the regression coefficient 0,241, followed by the promotion variable with a regression coefficient 0,208, price 0,018 and the variable have the negative influence is place with the regression coefficient -1,446. The computation of hypothesis using the t test showed that the variable product and promotion in meticulous proved significant. And the variable price and place not significantly influence the purchase decision variables. Then through the F test can be know that the independent variable is feasible to test purchasing dependent variable. Figures adjusted R square of 0,262 indicates that variable of purchasing decision can be explained by four independent variables in the regression quotation. The remaining 66,6% is explained by other variable out side of the three variables used in this research.